I don’t know what the difference between left and right, I know what the left is but cannot get the right

Hi, @ricardowu1112 !

Both graphs show roughly the same information. The only difference I can see is that in the left graph the output has been filtered by some sort of confidence threshold to get a binary “yes” or “no” rather than probabilities. I agree with you that it may have a bit clearer explanation.

So the left one is probability and the right one is label based on probability?

Yes, that’s what it seems

In addition to the difference mentioned by @alvaroramajo, a main feature on the left graph is the shaded area. A mesh grid is overlayed on the graph and the more “bluish” it is, the higher the probability value. It is there we see pale blue along the decision boundaries.

Cheers,

Raymond